ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stefan Bodewig <bode...@bost.de>
Subject Re: [RFE] Richer Task Specification
Date Tue, 27 Jun 2000 08:27:17 GMT
>>>>> "sr" == rubys  <rubys@us.ibm.com> writes:

 sr> Stefan Bodewig wrote: 

 sr> If we believe that users should be aware of the distinction
 sr> between javac-like things (tasks), and taskdef-like things
 sr> (declarations), then we shouldn't encourage people to develop bad
 sr> habits to begin with.

 >>  Agreed - you could remove the "if" part and we'd still agree
 >> though.

 sr> Actually, I don't see the need to keep people aware of such
 sr> distinctions.  The less that people have to learn to use this
 sr> tool, the better, IMHO.

No, I didn't want to imply we'd agree on "we believe that users should
be aware". I meant, you could just say

we shouldn't encourage people to develop bad habits to begin with.

and nobody is going to object.

I still feel there is a difference between tasks and declaration type
things. But I agree it's not really necessary to force this
distinction onto (new) users.

Let's see where we are: 

1. You've convinced me that tasks outside of targets are a good thing
with the "no targets at all" example. If I want this functionality
I'll need to allow for tasks outside of targets in the general case -
still don't like it but there is no other way taking the simplicity
goal into account.

2. I still want the depends attribute on taskdefs to allow users that
don't want tasks outside of targets to generate their taskdefs on the
fly.

Stefan

Mime
View raw message